King Kaufman's Sports Daily

How can Barry Bonds be the MVP when he's such a lousy left fielder? The Stat of the Day tells all. Hint: It's a trick question.

Published September 23, 2004 7:00PM (EDT)

Welcome to the penultimate Barry Bonds for MVP Stat of the Day. It's almost time to retire this feature and turn our attention to the always exciting struggle among mediocre teams for the last playoff spot.

At the moment in the National League the Giants are threatening to jump out of that contest against the Cubs and Astros by taking the lead in the West, but fear not. The Dodgers are playing poorly enough that they'll parachute right into the wild card race if the Giants pass them.

A gratifying number of readers, at least two or three, have written in to say the BB4MVPSotD has swayed them. "I was pretty much in favor of Barry winning the MVP before, but any ambivalence I had has been blown away," reads one from Tom Abernathy. "I give; you've convinced me." Jason Hare writes from St. Louis, anti-Bonds central: "This Cardinal fan is convinced."

But there are still objections to the idea of Bonds winning his seventh Most Valuable Player award. They are, in decreasing order of ridiculousness: He's won enough MVP awards already, he's a jerk, he's a horrendous left fielder and he's on steroids.

We'll talk steroids Friday.

A player can't win too many awards. If he deserves it every year, he should win it every year. If you don't like it, be patient. He'll get old and fade away eventually, though with Bonds, who's 40, one has to wonder. Having an award for Most Valuable Player Except for the Guy Who's Really the Most Valuable Player is just beyond stupid.

As for Bonds being a jerk: It's unwise to judge someone you've never met based on how he's portrayed in the media. A lot of people who come off as pretty nice are really pretty jerky, and vice versa. My own personal experience with Bonds is extremely limited, and I can't say he disabused me of the notion that he's a jerk. But I also can't say he had much of a chance or a reason to do so.

And I can also say that in limited personal interactions with the other MVP "candidates," I've found at least one of them to be a complete prick, which doesn't mean he really is one in some objective way, and which also doesn't alter my view of him as a ballplayer one tiny bit. If winning the MVP had anything to do with being a nice guy, Harold Reynolds would have won a truckload of them. Prince of a guy, from what I hear.

So that brings us to the Barry Bonds for MVP Stat of the Day: Defense.

OK, that's not a stat, but we're going to talk about it anyway.

The argument here is that Bonds doesn't deserve to be the MVP because he's a terrible left fielder, slow and lazy and with a rag arm. Those making this point correctly note that the award is Most Valuable Player, not just Most Valuable Hitter. All that offensive production by Bonds, the argument goes, is canceled out by his awful defense.

There are two problems with this argument. The first is that even if Bonds were the worst left fielder in memory, he probably couldn't do enough damage in the field to cancel out the offense he produces.

The second, and more important, is that Bonds isn't a terrible left fielder. He's not even a bad one.

Defense is the great frontier of statistical analysis in baseball. As much as stats can tell us about hitting and pitching, they lag badly in analyzing fielding. There are some useful stats, but they are much rougher than the ones that measure the relatively isolated transaction between pitcher and batter.

Fielding is a team event. There's teamwork involved and overlapping responsibilities. So it presents similar problems for the statistical analyst that football, basketball and hockey do.

If you're a right fielder who plays next to a rangy center fielder, he's going to come over and catch a lot of balls that in a normal outfield would be yours, and the stats are going to say you have limited range. If a shortstop makes a ton of plays, it might be because he has exceptional range, but it also might be because he plays behind a bunch of ground-ball pitchers and next to a slow third baseman.

An additional problem is that our eyes can deceive us. We tend to think of the best fielders as the ones who make the most spectacular plays, but how often is a diving circus catch the result of a fielder getting a poor jump or taking a bad route to the ball? A better fielder, one who knows where to position himself or gets a better jump, might make that same play look routine, and fail to impress us.

But like I say, there are some tools, and what they tell us in a nutshell is that Bonds is a perfectly adequate left fielder, a shade below league average. He's not the brilliant Gold Glover he was in his younger days, but he's hardly a liability.

This should be obvious without looking at any stats by considering how often Giants manager Felipe Alou has to agonize over whether to sacrifice potential offense by bringing in a defensive replacement for Bonds late in close games, which is never.

But let's look at some stats. I mentioned that Bonds is just below league average as a fielder, but league average includes all-field no-hit guys. There are eight National League left fielders who have hit enough and stayed healthy enough to play the position in two-thirds of their team's games. Aside from Bonds they are Moises Alou of the Cubs, Pat Burrell of the Phillies, Adam Dunn of the Reds, Cliff Floyd of the Mets, Luis Gonzalez of the Diamondbacks, Matt Holliday of the Rockies and Geoff Jenkins of the Brewers.

Bonds is in the top half of that group defensively.

Bonds is tied with Burrell for third among those eight in fielding percentage, the classic defensive metric. But since fielding percentage is based on errors, it rewards a guy for not even being good enough to get to a ball so he can boot it. That Burrell, easily the best regular left fielder in the league, is only tied for third tells you a little something about fielding percentage.

Two more common stats measure range. Range factor simply counts up how many plays a fielder makes per game, and zone rating defines a zone of responsibility for each fielder, then counts the percentage of balls in that zone he makes a play on. Bonds is third in range factor, fifth in zone rating among the eight regular left fielders.

I mentioned the Web site Baseball Prospectus the other day, and I want to reiterate what great work the folks there do in analyzing baseball, typically, but not always, from a statistical viewpoint.

They have various ways of measuring defense that are a little more complex and, I think, a lot more telling than the more traditional stats. Without getting too deeply into it, I'll just pick one, my favorite, which is simply called "rate." Here's the definition from the Web site: "A way to look at the fielder's rate of production, equal to 100 plus the number of runs above or below average this fielder is per 100 games. A player with a rate of 110 is 10 runs above average per 100 games, a player with an 87 is 13 runs below average per 100 games, etc."

I like to translate the rate from 100 games to 150, which is basically a full season for a fielder, figuring everybody gets a few days off. So if a player has a rate of 102, that means that every 100 games, he's saving his team two runs on defense against what a league-average fielder would do. That would be three runs for the full season.

Burrell, leading the league by a lot, has played left well enough to save his team six runs during the season over an average left fielder. Gonzalez has played left poorly enough to cost his team 13.5 runs that an average fielder would have prevented. (I couldn't find numbers for Holliday, a rookie, but the traditional stats suggest he's pretty lousy too.) So the difference between the best regular left fielder in the league and the worst is about 20 runs.

Bonds has a rate of 98, tied for third among the regular left fielders with Floyd, behind Burrell and Jenkins. So his defense costs his team three runs over a season against an average fielder. That's it. Three runs. It costs his team nine runs against what the best left fielder in the league would prevent. Nine runs.

Offensively, Bonds has produced 42 runs more than the second best hitter in the league, Jim Edmonds; 58 runs more than the second best hitting left fielder, Dunn; and 91 runs more than the average of the seven other regular left fielders in the league.

Of course, the other side of the Bonds-as-lousy-fielder argument is that the other MVP candidates, Edmonds, Scott Rolen and Albert Pujols of the Cardinals and Adrian Beltre of the Dodgers, are all fine fielders. Pujols is a pretty good first baseman, Rolen and Beltre at third and Edmonds in center are all superlative.

Based on rate, Rolen is 21 runs better than average, which is getting into Brooks Robinson/Clete Boyer territory. Edmonds is 17 runs better than average, Beltre 14 and Pujols six. Bonds, again, is three below average, so the biggest spread -- it's a little iffy to compare across positions like this but I'll do it anyway -- is the 24 defensive runs between Bonds and Rolen.

Considering that none of the others are within 40 runs of Bonds offensively, his defense doesn't disqualify him for MVP. It doesn't even make it a close race.

I don't know who the worst left fielder of all time was, but the worst I've ever seen was Greg Luzinski, who at his worst would cost the Phillies 25 runs an average left fielder would have prevented. If Bonds played left like Luzinski did, he'd still be more valuable than the other candidates, with the possible exception of Edmonds, who might just nip him.

And one more traditional stat: Bonds leads National League left fielders in assists with 11. This doesn't prove anything but it does suggest that opposing players and third-base coaches share the belief that Bonds is lousy in left, and that this belief is costly.

Previous column: Johan Santana

- - - - - - - - - - - -

  • Bookmark to get the new Kaufman column every day.
  • Discuss this column and the sports news of the day in Table Talk.
  • Send an e-mail to King Kaufman.
  • To receive the Sports Daily Newsletter, send an e-mail to

  • By Salon Staff

    MORE FROM Salon Staff

    Related Topics ------------------------------------------